Study of the antidepressant mechanism of electroconvulsive therapy (ECT) is of particular importance because of its: efficacy; uniqueness as a nonchemical centrally-directed treatment; and lack of pharmacokinetic confounding effects. It has been hypothesized on the basis of animal studies showing beta receptor downregulation and 5-HT2 receptor changes (downregulation by antidepressant drugs and upregulation by ECT) that these effects may mediate the antidepressant effect of ECT. However, no such receptor studies have been reported to date in ECT-treated depressed patients. Moreover, we have found an increase in platelet 5-HT2 binding sites and subsensitive beta adrenergic receptor function in depressed patients. The concept that the receptor systems through which ECT exerts its antidepressant effect are the same ones that are altered in depression is an exciting possibility. We will study: 1) lymphocyte beta receptor binding; 2) isoproterenol-sensitive adenylate cyclase response; 3) hormonal regulation of beta receptors by catecholamines and cortisol; 4) acute responses of catecholamines, heart rate, and blood pressure to ECT; 5) 24 hour urinary excretion levels of MHPG and 5-HIAA; 6) platelet 5-HT2 binding; and 7) rise in plasma growth hormone, prolactin, and cortisol after fenfluramine challenge as a measure of central 5-HT function. Patients with a major depressive episode, endogenous subtype will be compared at baseline to controls, and then studied longitudinally during the course of ECT. They will be studied 24 hours after: ECT #1 (effect of 1 ECT); ECT #5 (to distinguish effects specifically correlated with degree of antidepressant response); and after the final ECT. Enduring effects will be assessed by studying patients 5-7 days after the last ECT. These measures may not only have potential as diagnostic tools (state versus trait markers), and monitors of clinical response, but may lead to better understanding of mechanism of action of ECT and thereby to improved chemical antidepressants.